Abstract
needs a solid validation that has verification and matching uploaded images. To solve this problem, this paper implementing a detection model using Faster R-CNN and a matching method using ORB (Oriented FAST and Rotated BRIEF) and KNN-BFM (K-Nearest Neighbor Brute Force Matcher). The goal of the implementations is to reach both an 80% mark of accuracy and prove matching using ORB only can be a replaced OCR technique. The implementation accuracy results in the detection model reach mAP (Mean Average Precision) of 94%. But, the matching process only achieves an accuracy of 43,46%. The matching process using only image feature matching underperforms the previous OCR technique but improves processing time from 4510ms to 60m). Image matching accuracy has proven to increase by using a high-quality dan high quantity dataset, extracting features on the important area of EKTP card images.
Highlights
An EKTP image repository can be a helpful tool to assist human operators in EKTP image pair checking
In 2013, 64.5% of that business process started to accept EKTP recorded in Palembang [4]
In Cibeuying Kaler recorded from 100 citizen respondents to an EKTP survey satisfaction, 84,9% of the respondent is not image of a person holding an EKTP card of theirs
Summary
In Indonesia, using a national identity card called EKTP (Kartu Tanda Penduduk Elektronik) is very substantial. The result of implementation will be evaluated based carried out by Faster R-CNN, ORB, and KNN-BFM [9]- on the total feature matched in the data image pair. E-KTP data collection by EKTP detection and image matching process will be Indonesia citizen identity card or E-KTP is compulsory repository validation for future uploaded images. An illustration of Faster R-CNN Architecture obtained from making binary comparisons of 2 randomly selected pixel points This process can be Machine learning model accuracy is directly calculated using the following formula (3). Weight of pre-trained Faster R-CNN architecture with the same base model as the original paper proposed [9] This ORB technique is used as feature extraction from Average Precision calculates the maximum precision the segmented image to match the identity document value for each recall value on the N available data in the card image input. Performance in various test cases [13], which is expected to solve the problem domain of this study
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have